AT-FAULT CMV CRASHES: Top Behavioral Factors And Outcomes

Transcription

1 TEXAS TRUCKING ASSOCIATION SAFETY MANAGEMENT COUNCIL Spring Seminar March 22, 2018 Waco, Texas AT-FAULT CMV CRASHES: Top Behavioral Factors And Outcomes Analysis from 20 Texas Counties,

2 Project background/goals Combine safety data Evaluate top crash factors Compare oilfield & non-oilfield locations Visualize CMV crash risk Inform safety efforts Funding through FMCSA, TTI, and the Center for Transportation Safety Nixon, Texas, ca Image by D. Bierling, Texas A&M Transportation Institute.

3 What did project include? 2,780 CMV crashes Controlled for type, configuration, weight At-fault crashes only 20 counties across Texas Years Polk County, Texas, ca Image by Texas A&M Transportation Institute

4 Sample area Counties geographically distributed across Texas

5 County selection 10 oilfield counties 10 non-oilfield counties Matched by percentiles: VMT Lane Miles Population Area ADT Pop. density Also, network configuration

6 Data sources... Road-Highway Inventory Network (RHiNo), TXDOT Crash Record Information System (CRIS), TXDOT Commercial vehicle inspection database, TXDPS CVE Well location data, Texas Railroad Commission (RRC) Drilling permit data, Texas RRC American Community Survey, U.S. Census Bureau

7 Excluded crashes Not on state road network No CMV at-fault More than one CMV at-fault Crashes between CMV and only non-motor vehicles Crashes involving 9+ passenger vans, school buses, commercial buses, or bus body styles Railroad-related crashes Crashes where at-fault CMV was parked All other at-fault CMV crashes on state roadways in sample counties were included: 2,780 crashes.

8 Crash severity measures Based on reported injuries/fatalities Estimated Crash Cost Using National Safety Council economic impact values KABC/O CMV and passenger vehicle crash on I-10, Sutton County, ca Image by D. Bierling, Texas A&M Transportation Institute.

9 Stacked data 1. Physical context Location, geographic setting 2. Roadway properties Infrastructure, management, administration 3. Driving environment 4. Driving apparatus Vehicle characteristics 5. Driver characteristics 6. Driver behaviors 7. Crash parameters

10 Driver behaviors Variable Group Contributing factors Citations Variable Backing-related Parking-related Passing-related Lane use-related Stopping-related Turning-related Distracted driving-related Speed-related Failure to yield right-of-way (FTYROW) Driver fatigued, ill, or impaired Dangerous driving-related Load securement or size-related Other moving contributing factors Vehicle defects-related Alcohol or drugs Unsafe backing Disregarded sign or signal Driving where prohibited FTYROW - Driveway, intersection, stop sign, or signal FTYROW - Turning-related FTYROW - Other Intoxicated assault Load securement, size, or weight Other moving violations Passing or lane use-related Speed-related Vehicle or equipment defects Contributing factors, vehicle defects, and citations from crash reports Grouped by frequency and type/similarity Vehicle defects grouped in one category (low individual frequencies) Also used for at-fault assignment Source: Polk County Today (

11

12 Discussion: Contributing SPEEDING factors was a factor in 20% of intersection crashes and increased estimated crash costs by 20%. When a SPEEDING citation was issued in at-fault CMV crashes, there was a 170% increase in risk of injuries & fatalities. IMPROPER LANE USE was a factor in 31% of single vehicle crashes, resulting in more than 30% higher injury/fatality risk per crash.

13

14

15 Oilfield effects (or not) Oilfield/non-oilfield measures did not retain significant relationships with at-fault CMV crash severity, accounting for other variables in models Unexpected result Crash frequencies in oilfield areas are higher, but not severity per crash

16 Other factors Rural crashes more severe, even after accounting for higher speed limits Interstate/freeway versus arterial crashes have different relationships for intersection/non-intersection The following were also associated with increased at-fault CMV crash severity (depending on model): Lower socio-economic status of crash location Low/intermediate flex road surface (compared to high flex road surface); load-limited roads Van cargo body style (compared to other styles) Heavier (5+ ton) CMV Dark or nighttime crashes (compared to light/day) Intrastate carrier (compared to interstate)

17 Application Mapping examples Atascosa County, Texas

18 For discussion Which crash factors (driver fatigue, stopping problems, lane use, speeding, overturns/rollovers) are most important for TXTA SMC members? Which crash factors are easiest to change? How are cost/risk data used to guide training? Are cost and injury fatality risk rates higher or lower than expected? How are maps that visualize crash locations, causes, and route risk profiles best used by carriers? By operators?

19 Integrating Datasets to Evaluate CMV Crash Severity and Risk on Texas Roads Project Team: David Bierling, Boya Dai, Srinivas Geedipally, Katrina Korenek, Michael Martin, Zachary Palisch, Ashesh Pant, Marcie Perez, Sushant Sharma, Eva Shipp, Amber Trueblood, Ioannis Tsapakis Project Supervisor: David Bierling, Ph.D. Research Scientist Texas A&M Transportation Institute SH 72, Tilden, Texas, ca Image by Debbie Jasek, Texas A&M Transportation Institute